Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utcsri!utegc!utai!gh
From: gh@utai.UUCP
Newsgroups: soc.college,comp.edu
Subject: Re: Preparing for GRE's (especially Computer Science)
Message-ID: <3272@utai.UUCP>
Date: Fri, 13-Feb-87 12:14:21 EST
Article-I.D.: utai.3272
Posted: Fri Feb 13 12:14:21 1987
Date-Received: Fri, 13-Feb-87 23:24:45 EST
References: <9186@duke.duke.UUCP> <1752@cit-vax.Caltech.Edu>
Reply-To: gh@utai.UUCP (Graeme Hirst)
Organization: Dept of Computer Science, University of Toronto
Lines: 31
Summary: GREs do matter (like it or not), but aptitude counts more.

In article <9186@duke.duke.UUCP> ravi@duke.UUCP (Ravi Subrahmanyan) asks:
>Do GREs really matter?

In article <1752@cit-vax.Caltech.Edu> jon@oddhack.UUCP (Jon Leech) answers:
>
>	Depends on the grad school. I found it trivial to get an impressive
>score but some people seem to think GREs actually indicate something other
>than one's ability to memorize.

The fact is that most admissions committees look at GREs pretty closely,
because, whatever their flaws, they are the only measure that is standardized
across schools.  How else can you decide between someone who got B grades at
Stanford and someone who got A+ grades at South Dakota State?

When I was on such a committee [at Brown University], we looked very hard at
the general aptitude scores (which do NOT reflect any memorization ability,
though they clearly reflect a general test-taking ability, and I strongly
recommend taking the practice tests available).  The subject GRE only mattered
if it was really bad.  For us, you needed to be well into the ninety
percentiles in aptitude, but eighties or high seventies in computer science was
okay.  In fact, most people who got 95 percentile or better in the aptitude
tests did not do so well in the Computer Science test, and I started to wonder
if anyone actually scored in the top ten percent in the test!

Perhaps the good students are exactly the ones who understand instead of
memorize and hence get respectable but not outstanding scores in the subject
test, while the memorizers do well in the subject test but don't make it in the
aptitude tests.
-- 
\\\\   Graeme Hirst    University of Toronto	Computer Science Department
////   utcsri!utai!gh  /  gh@ai.toronto.edu  /	416-978-8747